=Paper=
{{Paper
|id=Vol-1964/S2
|storemode=property
|title=Continuous Authentication on Smartphones Using An Artificial Immune System
|pdfUrl=https://ceur-ws.org/Vol-1964/S2.pdf
|volume=Vol-1964
|authors=Nawaf Aljohani,Joseph Shelton,Kaushik Roy
|dblpUrl=https://dblp.org/rec/conf/maics/AljohaniS017
}}
==Continuous Authentication on Smartphones Using An Artificial Immune System==
https://ceur-ws.org/Vol-1964/S2.pdf
Nawaf Aljohani et al. MAICS 2017 pp. 171–174
Continuous Authentication on Smartphones Using
An Artificial Immune System
Nawaf Aljohani1, Joseph Shelton, Kaushik Roy
Department of Computer Science, North Carolina A&T State University, Greensboro, U.S.A
naaljoha@aggies.ncat.edu1
Abstract [Chang et al., 2012]. Nowadays, most mobile devices use
Most of the authentication systems require the users to graphical password that have a larger and more accepted
provide their credential for authentication purposes by password space. Though graphical password increases the
providing their passwords or their biometric data. password space in touch screen handheld mobile devices,
However, as long as the user remains active in the there are no further authentication processes after
system, there is no mechanisms to verify whether the
user who provides the credential is still in control of unlocking the touch screen. Thus, the attacker has the
the device or not. Most mobile devices rely upon ability to access and control all the users’ data and
passwords and physical biometrics to authenticate resources as long as the attacker gains access to a device
users only when they start using the device. Active after it is unlocked. This research aims to continuously
authentication based on analyzing the user’s touch authenticate the users without asking them to provide the
interaction could be a reasonable solution to verify
that a legitimate user is still in control of a smartphone login information multiple times while the smartphones or
or tablet. In this research, an Artificial Immune tablets are in use.
System (AIS) is proposed to apply to continuously
authenticate the users based on touch patterns. Our In this research, an artificial immune system (AIS)
results show that AIS is able to actively authenticate approach will be used to secure mobile devices. The
96.89% of the users correctly.
immune system is considered to be a highly complex
functional system that protects the body from foreign
Introduction diseases causing pathogens [Shojaie and Moradi, 2008].
During the authentication process, a primary concern for This immunology inspired researchers to develop the
computational intelligence technique, which is called AIS.
users and designers is the level of security. The process of
AIS has been used in solving complex computational
authenticating an individual must be both secure and
problems, such as classification, recognition, and network
effective to be applicable for a real world authentication security [Dudek, 2012]. This research makes use of an AIS
system. In the event that the authentication process is which has the ability to continuously keep track of any
compromised, other aspects in the system such as changes in the environment based on recognizing the
availability, confidentiality, and integrity would be easily patterns of ‘self’ and predicting and detecting new patterns
compromised as well. Knowledge-based authentication of ‘non-self’.
systems, such as password or pin, have several drawbacks,
but many systems still use this method to authenticate This research uses a set of 11 behavioral touch features that
legitimate users due to their simplicity and flexibility. This were extracted while the users were interacting with their
smartphones. This research uses touch data collected from
research proposes an authentication method for the users
100 users and each subject has 100 instances [Sitová et al.
based on finger swipe movements. 2016]. This research proposes the use of an AIS to
continuously authenticate the smartphones users where the
Touch screen technology is used in many mobile devices
security of smartphone is enhanced.
where users have the ability to access various data and
resources at anytime. Most of the smartphones use PINs to
authenticate the users. However, a traditional PIN typically Related Work
consists of four to eight digits, making it easy to guess with Sitová et al. proposed a set of behavioral features based on
its small password space and thus vulnerable to attacks hand movement, orientation, and grasp to continuously
authenticate mobile users [Sitová et al. 2016]. The data is
Copyright held by the author(s). collected from 100 participants under two conditions:
171
Continuous Authentication on Smartphones Using An Artificial Immune System pp. 171–174
walking and sitting. The achieved equal error rates (EERs) Proposed Approach
are 7.16% (walking) and 10.05% (sitting) where walking
interactions are more richer than sitting interactions due to There are three types of AISs reported in the literature: 1)
the distinctive body movements caused by walking and Negative selection, 2) Clonal selection, and 3) Immune
hand-movement dynamics from taps. Sitová et al. believe Network [Watkins et al., 2002].
that each mobile user has postural preferences for
interacting with touch screen which can be used to The main goal of the Negative Selection (NS) is to provide
authenticate the users. In their dataset, Sitová et al. tolerance for self cells that indicate the ability to detect non
designed 96 features and extracted data while users are self antigens. This idea is used in many areas such as
walking and sitting. The dataset was divided into two types network security, where NS generates detectors and then
of features that grasp resistance and stability and the data removes those that can detect self patterns. The rest of
was collected by using three sensors: accelerometer, detectors can be used to detect anomaly. The detectors,
gyroscope, and magnetometer [Sitová et al. 2016]. The which are randomly generated, are representation for
user touch data was acquired using Samsung Galaxy S4 matching the authorized users’ patterns to create the self
smartphone where the average duration of a user’s profile [Greensmith et al., 2010]. A detector is a set of
interaction with touch screen was 11.6 minutes per session. intervals for each feature a detector is created for. Any self
Sitová et al. used scaled Manhattan (SM), scaled Euclidian pattern number lies in the interval of a detector means that
(SE), and 1-class Support Vector Machines (SVM) to the detector detects the self pattern. As a result of that all
verify the users. detectors that detect self patterns must be removed. The
remaining detectors are used to detect unauthorized users.
Frank et al. [Frank et al., 2013] determined whether or not
a classifier could be used to continuously authenticate Clonal Selection (CS) differs from the NS approach by
users based on their interaction with the touch screen. selecting the detectors that proliferate over those that do
Authors in [Frank et al., 2013] proposed 30 behavioral not. The main feature of CS is the new detectors that are
touch features that could be collected from raw touch the copies of their parents and reactivated detectors are
screen logs. These features were used to identify a user eliminated afterwards. In this research, CS is implemented
based on the way he/she interacts with the touch screen. to authenticate smartphone users continuously instead of
Furthermore, Frank et al. explained the reasons that mobile NS because CS, in some cases, gives better accuracy than
devices are at higher risk than that of desktop computers NS due to the fact that the nearest created detectors cloned
due to that fact that mobile devices can be easily lost or (See Figure 1). First, CS generates n detectors and searches
stolen. Their dataset consists of 41 subjects and the data for new patterns. CS selects the nearest detectors to be
was collected from four different smart phones. cloned using distance metric such as the Euclidean
distance. CS clones the nearest detectors from the detectors
Feng et al. [Feng et al., 2012] introduced FAST: and the new pattern. CS then finds best matching clone
Fingergestures Authentication System using Touchscreen. and assigns clone class to antigen. Finally, it deletes other
Their idea was to extract data from touch screen superfluous clones and for each deletion, replaces with
interactions and validate the data by using a digital sensor new randomly generated detectors [Greensmith et al.,
glove. Their proposed approach relied on Random Forest 2010]. In this research, we created 100 detectors and then
(RF) and Bayesian network classifiers to authenticate remove those detectors that can detect self elements. Self
mobile users continuously. Feng et al. used a dataset that elements in the dataset represent one subject. The
consisted of 40 users and authors obtained a False Accept remaining subjects are non-self elements. The detectors
Rate (FAR) of 4.66%, while the False Reject Rate (FRR) detect the self elements by exploring self subject’s features
was 0.13%. patterns. Suppose the smartphone is unlocked by an
attacker, detectors created by CS are used for continuously
Meng et al. [Meng et al., 2013] proposed a scheme based authenticating the user by tracking the user’s interactions
on touch dynamics that used a set of behavioral features to with touch screen. As long as the smartphone is unlocked,
improve the accuracy of user authentication. Their dataset CS detectors are used to analyze the user’s interactions via
detectors. Once a certain number of detectors detect
consisted of 21 features that were collected form users
abnormal interactions, the device is locked due to
interaction with the touch screen. All data was extracted
unauthorized access. In our experiments, abnormal
from 20 Android phones. Researchers in [Meng et al.,
interactions must be detected by four detectors to detect
2013] found that a Neural Network (NN) achieved an unauthorized access.
average error rate of 7.8%. In addition, Meng et al.
optimized the NN by implementing Particle Swarm
Optimization (PSO) and they reported an average error rate
of about 3%.
172
Nawaf Aljohani et al. MAICS 2017 pp. 171–174
We ran the AIS on the entire dataset for 20 times. For each
run, we experimented it for 1000 generations. Also, for
each run, we use 100 detectors. The best accuracy for 20
runs is 96.89% and the average is 93.81% (See Figure 3).
The average of FRR out of 20 runs is 0.9381 which shows
unauthorized users detections. The average of FARs on the
other hand is 0.06.
Figure 1. Clonal selection concept
Table 2. Results of adding each user
User Accuracy User accuracy User accuracy
Table 1. behavioral features [Sitová et al., 2016]. 2 1 35 0.939246 68 0.919757
Name Description 3 1 36 0.9409309 69 0.924135
Systime Absolute time-stamp 4 1 37 0.9417141 70 0.920664
5 1 38 0.9473414 71 0.922134
EventTime Sensor event relative time-
6 0.993333 39 0.9471731 72 0.927492
stamp
7 0.993036 40 0.9375915 73 0.923671
ActivityID Belonged activity 8 0.991111 41 0.938403 74 0.923728
Pointer_count 1: Single touch 9 0.991444 42 0.941495 75 0.924533
2: Multi-touch 10 0.994636 43 0.940016 76 0.919508
PointerID 0: Single touch; or first pointer 11 0.99447 44 0.941293 77 0.927949
in multi-touch 12 0.991282 45 0.952014 78 0.930393
1: Second pointer in multi- 13 0.987418 46 0.940592 79 0.927764
touch 14 0.970905 47 0.941241 80 0.923218
ActionID 0 or 5: DOWN 15 0.9675 48 0.941101 81 0.92879
1 or 6: UP 16 0.964118 49 0.945453 82 0.922505
2: MOVE 17 0.966471 50 0.944878 83 0.926952
X Touch location in X 18 0.965409 51 0.944246 84 0.897615
coordination 19 0.964263 52 0.939213 85 0.926551
20 0.950119 53 0.941237 86 0.927851
Y Touch location in Y
21 0.963009 54 0.946933 87 0.921804
coordination
22 0.9543478 55 0.936571 88 0.925723
Pressure Touch pressure 23 0.9558696 56 0.946563 89 0.923815
Contact_size Touch contact size 24 0.9473167 57 0.939958 90 0.926972
Phone_orientation 0: Portrait and no rotate 25 0.9522769 58 0.955611 91 0.923371
1: device rotated 90 degrees 26 0.9377635 59 0.954308 92 0.917491
counterclockwise 27 0.9335714 60 0.931101 93 0.91647
3: device rotated 90 degrees 28 0.9355788 61 0.913392 94 0.910208
clockwise 29 0.9428851 62 0.919519 95 0.917088
30 0.9464194 63 0.914184 96 0.921645
31 0.9375403 64 0.916935 97 0.919436
32 0.9340909 65 0.922818 98 0.925315
Experimental Results 33 0.9328699 66 0.921142 99 0.922192
We conducted our experiments on TouchEvent dataset that 34 0.9395126 67 0.922046
has 11 behavioral features [Sitová et al. 2016]. A list of
features is shown in Table 1. First, we ran the AIS with
only two subjects, and we achieved an accuracy of 100%.
The accuracy remains 100% as long as we are using 5
subjects. After adding 6th subject and its instances, the
accuracy went down to 99.33%. Initially, we added a new
subject to the dataset and ran AIS. The accuracies for all
the users are shown in Table 2.
It is clear that at a certain point, adding users do not affect
the accuracy of AIS as shown in Figure 2. For each run,
AIS is executed for 1000 generations and the number of
detectors is 100. Unauthorized user is detected if at least 4 Figure 2. CS Performance
generated detectors are able to detect the touch screen
interactions.
173
Continuous Authentication on Smartphones Using An Artificial Immune System pp. 171–174
[Frank et al., 2013] Frank, M., R. Biedert, E. Ma, I. Martinovic,
and D. Song. (2013) "Touchalytics: On the Applicability of
Touchscreen Input as a Behavioral Biometric for Continuous
Authentication." IEEE Trans.Inform.Forensic Secur. IEEE
Transactions on Information Forensics and Security 8.1 pp.136-
48. Print
[Greensmith et al., 2010] Greensmith, J., Whitbrook, A., &
Aickelin, U. (2010). Artificial immune systems. In Handbook of
Metaheuristics (pp. 421-448). Springer US.
[Meng et al., 2013] Meng, Yuxin, Duncan S. Wong, Roman
Schlegel, and Lam-For Kwok. (2013) "Touch Gestures Based
Figure 3. CS Performance
Biometric Authentication Scheme for Touchscreen Mobile
Phones." Information.Security and Cryptology Lecture Notes in
Computer Science. pp. 331-50. Print.
Conclusion and Future Work [Shojaie et al., 2008] Shojaie, S., & Moradi, M. H. (2008). An
evolutionary artificial immune system for feature selection and
We find from the experimental results that AIS can be used parameters optimization of support vector machines for ERP
to authenticate smartphone users continuously. AIS assessment in a P300-based GKT. In Biomedical Engineering
approach has the flexibility in restricting the authentication Conference, 2008. CIBEC 2008. Cairo International (pp. 1-5).
process based on the sensitivity of the mobile devices by IEEE.
reducing the number of detectors. The best performance of [Sitová et al., 2016] Sitová, Z., Šeděnka, J., Yang, Q., Peng, G.,
CS on the entire dataset was 96.89%. However, there Zhou, G., Gasti, P., & Balagani, K. S. (2016). HMOG: New
behavioral biometric features for continuous authentication of
seems to be a promise with the increasing number of smartphone users. IEEE Transactions on Information Forensics
detectors for each run. This research uses CS for an AIS to and Security, 11(5), 877-892.
continuously authenticate the users.
[Watkins et al., 2002] Watkins, Andrew and Timmis,
Future work will be focused on evaluating the impact of Jon (2002) Artificial Immune Recognition System
each behavioral feature on the overall accuracy. (AIRS):Revisions and Refinements
Furthermore, future work will evaluate the effect of
increasing the number of detectors. This may improve
accuracy by increasing the chance of detecting more
unauthorized mobile interactions. Also, a comparison
between CS and NS will be conducted. In addition, the
performance of CS and NS will be compared with other
classifies such as SVM.
Acknowledgements
This research is supported by the Army Research Office
(Contract No. W911NF-15-1-0524).
References
[Chang et al., 2012] Chang, T. Y., Tsai, C. J., & Lin, J. H. (2012).
A graphical-based password keystroke dynamic authentication
system for touch screen handheld mobile devices. Journal of
Systems and Software, 85(5), 1157-1165.
[Dudek, 2012] Dudek, G. (2012). An artificial immune system for
classification with local feature selection. IEEE Transactions on
Evolutionary Computation, 16(6),p.847-860.
[Feng et al., 2012] Feng, Tao, Ziyi Liu, Kyeong-An Kwon,
Weidong Shi, Bogdan Carbunar, Yifei Jiang, and Nhung Nguyen.
(2012). "Continuous Mobile Authentication Using Touchscreen
Gestures." 2012 IEEE Conference on Technologies for
Homeland Security (HST). Print.
174